Good Research Practice What Is It?

(1)

Good ReseaRch PRactice – What is it?

1:2 00 6 G o o d R es ea Rc h P Ra ct ic e – W h at is it ? Ve te N sK a Ps RÅ d et s R a PP

Regeringsgatan 56 103 78 stockholm tel +_{46-8-546 44 000 Fax}+_{46-8-546 44 180 [email protected] www.vr.se}

the swedish Research council is a government agency funding basic research of the

Good Research Practice – What is it? explores some of the ethical issues that arise in research, and is intended to provide a basis for reflection and discussion. it is aimed at researchers in every field of inquiry, and in particular at postgraduate students and their supervisors. the three authors, Professors Bengt Gustafsson, Göran hermerén and Bo Petersson, are them-selves experienced researchers. they discuss the basic ethical principles applied in research. they also describe the existing regulatory framework in sweden, and consider the standards of conduct which apply, or should apply, among researchers and between researchers and others.

(2)

Good ReseaRch PRactice

– What is it?

Views, guidelines and examples

(3)

Good ReseaRch PRactice – What is it? Views, guidelines and examples

this report can be ordered at www.vr.se

Vetenskapsrådet

(the swedish Research council) 103 78 stockholm

sweden

Graphic design: erik hagbard couchér, Vetenskapsrådet english translation: Martin Naylor

cover illustration: digital Vision/simon du Buisson other illustrations: Robert Nyberg

(4)

PReface

Under its terms of reference, one of the functions of the Swedish Research Council is to “take initiatives to ensure that ethical issues receive attention in research and to disseminate information on such issues”.

Ethics is not about statutory rules and regulations. Although certain eth-ical questions do have to be formally regulated, ethics is primarily a matter of developing, encouraging and keeping alive an awareness and a discussion about appropriate standards of conduct. Given the long-term significance of research and its place in society, ethical issues are of particular import-ance in this field. It is therefore vital to encourage a debate on ethics, and to provide basic material for such a debate.

This publication is designed to do just that. It is not a rule book or a manual – although it does contain frequent references to rules and guide-lines. Primarily it is intended as a contribution to a debate, providing a basis for reflection and discussion. It is the work of three experienced researchers, Professors Bengt Gustafsson, Göran Hermerén and Bo Petersson. The authors have discussed their ideas and thinking on an ongoing basis as members of the Ethics Committee of the Swedish Research Council.

Some readers may well feel that much of what is written in this publica-tion is self-evident. And that is of course true – the ethical guidelines set out here are not something new, produced in isolation; to be effective, they need to rest on values that are widely embraced in our culture. It thus fol-lows that such guidelines, and the ways in which they are to be applied in different fields and different situations, have to be constantly discussed and tested; they must be developed as research and society develop.

A succession of highly publicized conflicts over questions of research ethics in recent years have also helped to underscore the necessity of lines. History shows, however, that it is not enough simply to issue guide-lines. We also have to keep them alive by discussing them, referring to them in education and training, and continuously revising them. There needs to be an ongoing debate about how they are to be applied, and that debate must have consequences for our practical actions.

We therefore welcome this publication, and hope that it will be read, discussed and used in education and training, in the development of local procedures and in a wide range of other contexts in which ethical aspects of research are important. Stockholm, January 2005

Bengt Westerberg Pär omling

(5)

1 iNtRoductioN

Good Research Practice – What Is It? is aimed at researchers in every field of inquiry, and in particular at postgraduate students and their supervisors. Its authors, Bengt Gustafsson, Göran Hermerén and Bo Petersson (who has also acted as editor), are or used to be members of the Swedish Research Council’s Ethics Committee, at whose meetings fruitful discussions of earl-ier versions of the text have taken place.

The overall aim of this publication is to discuss what can be considered to constitute good research practice. Society at large has a significant interest in such a discussion taking place, and among researchers themselves there is a clearly expressed need for deliberation on these issues, for one thing be-cause research and the environments in which it is conducted are constantly changing. Once primarily the work of individual scholars, research is now often undertaken by groups of investigators, or by several groups working on a collaborative basis. The funding required for research was often auto-matically available to those holding research posts, whereas nowadays it is provided for specific projects, often from external sources. This raises new questions regarding researchers’ responsibilities and the factors they need to take into account.

In this report, we present material at a number of different levels. On the one hand we describe existing systems of rules, on the other we wish to encourage a more basic discussion about the ethical principles that are applied in research. Another aim is to offer a starting point and basis for the discussion and development of good local procedures. Last but not least, we wish to engage in a more personal discussion about the standards of conduct that apply, or should apply, among researchers and between researchers and others.

Good research practice should promote high-quality research and, partly with that end in view, both foster good relations among researchers and between researchers and the general public, and counteract and prevent re-search misconduct. Rere-searchers should not, however, shy away from such conflicts as may arise when their results challenge widely held ideas.

Good research practice is not a matter of slavishly following a tradition, for example in terms of method or basic perspective, or doing what others do, or always sticking to safe, conventional ground. On the contrary, one characteristic of good research practice is that the researcher helps to gen-erate new results, which often means departing from or questioning in-grained ways of thinking or working.

(9)

Faced with a choice between a concise, outline survey and a more detailed ac-count, we have by and large opted for the former. Readers wishing to go into the subject in greater depth will therefore need to supplement this report with other literature. One starting point for our work has been the Swedish

Research Council’s Riktlinjer för god medicinsk forskning (“Guidelines on good

medical research”, originally published by the former Swedish Medical Re-search Council, MFR). A number of wordings have been borrowed from that publication. Other basic material has been provided by various international documents in this field, some of which are presented in chapter 9.

Good Research Practice – What Is It? is not a treatise in the theory of sci-ence, nor does it represent a clearly defined position in that regard. Since it is written for researchers working at all levels and in all fields of inquiry, this is unavoidable. In the bibliography, however, we include a number of works on the theory and philosophy of science, for those wishing to pursue such issues further.

The focus of attention in this volume is on the conduct of the researcher in relation to the standards normally directly associated with science. Other aspects of the researcher’s work, such as ethical issues relating to the inter-ests of human subjects, are touched on to only a limited extent. In section 2.8 below, we attempt to distinguish between “research ethics” and what could be termed “the professional ethics of the researcher”. The present publication is chiefly concerned with the latter. It discusses what rights and obligations those working in research in various ways can reasonably be considered to have.

From a brief perusal of the recommendations set out in the following chap-ters it becomes clear that much of what is said there can be summed up in a number of general rules, all of which correspond to broader rules for life: 1 Tell the truth about your research

2 Openly report your methods and results

3 Openly disclose any commercial interests and other ties

4 Consciously examine and present the basic assumptions underlying your studies

5 Do not steal research results from others (e.g. from younger colleagues) 6 Conduct your research in an orderly manner (e.g. by maintaining

docu-mentation and retaining data)

7 Do not conduct your research in a way that could harm other people (e.g. subjects)

8 Be fair in your assessment of other people’s research

These requirements can be summed up in words like Honesty (1, 5),

Open-ness (2–4), Orderliness (6), Consideration (7) and Impartiality (8). Although

(10)

the rules enumerated here may seem general and applicable to all areas of research, and perhaps appear self-evident, their relative importance is not self-evident, least of all when conflicts arise between them. Furthermore, the significance we attach to them may depend on our view of science. Many, for example, work within the positivist tradition, which still holds a very strong position in the natural sciences and medicine, and which sees science as an ongoing accumulation of knowledge from one generation to the next. With this view of research results as something of great enduring value (provided that principles of reproducibility etc. are adhered to), the requirements of honesty and orderliness are natural in any endeavour to promote the growth of knowledge. As the recommendations in these pages make clear, the requirement of honesty is in fact considerably more stringent in research than it is in everyday life. White lies, say to protect a friend and colleague who happens to have been wrong about a scientific issue, are scarcely acceptable. And order as such is particularly important, as the accumulation of knowledge is viewed as a process that continues for generation after generation and involves the integration of reliable results from different studies into an ever growing body of knowledge. Such an at-titude need not entail a naive belief that all researchers produce objective truth. Precisely to be able to distinguish solid results, which can be built on, from shaky ones, the principles of honesty, openness and orderliness carry particular weight in this tradition.

If in the spirit of Thomas Kuhn, on the other hand, we believe that sci-ence develops through major revolutionary advances, or “paradigm shifts”, involving dramatic changes in fundamental outlooks, methods of meas- urement, unwritten rules and values, and our concern is to pave the way for such advances, then we can put particular emphasis on a critical scrutiny of our own basic assumptions, and on a criticism of the foundations of other people’s work. In practice, this is not easy. Kuhn argues, for example, that researchers’ basic assumptions and outlooks are largely unconscious or, for other reasons, not easily questioned by the researchers themselves. But that does not mean that conscious testing of those assumptions is not worth- while. The same is true if we regard research essentially as a cultural phenomenon, with cultural rules and outlooks more or less unconsciously pervading both research itself and its results.

Many, by contrast, share Karl Popper’s view that scientific advances ad-mittedly occur through rational and ordered processes, but are threatened by irrationality and elitism, with the result that science can easily become an instrument of power. With such a view, particular stress may need to be put on the requirements of openness and impartiality. Research is in danger of being subordinated to or incorporated into the power structures of

(11)

ciety, and could draw even its own practitioners into an illegitimate exercise of power. This is where the principle of consideration becomes important: research, like politics, business and other activities in society, must be subordinated to respect for human rights. What is more, research and researchers have important roles to play in society, and it is thus reasonable to place additional demands on the researcher. The question is how restrictive those demands should be.

In the present publication we have attempted to formulate a number of practical demands which we ourselves consider reasonable, and which we believe researchers should be able to accept without it posing too great a threat to the freedom of their work. It is our conviction that it is in the long-term interests of the research community, too, to respect the requirements and expectations placed on it in a democratic society.

The structure of this report is as follows. In chapter 2 we discuss the

over-all meaning of the concept of good research practice. The following chap-ters deal with the main task of the researcher: to do research. These sections

follow the development of a research project from planning (chapter 3), via

implementation (chapter 4) to publication (chapter 5), and we attempt to

formulate what good research practice requires in these contexts. Chapter 6

addresses the question of research collaboration, and examines good re-search practice in relation to other individuals or groups interacting with the researcher. This chapter also looks at different interests which affect

re-search and which may for example influence its direction. Chapter 7

consi-ders other roles of the researcher, and the ethical standards that apply when he or she undertakes, for example, the role of supervisor, teacher,

communi-cator, or member of a committee or board. Chapter 8 discusses the concept

of research misconduct. It also examines how instances of misconduct are to

be handled in practical terms, and how we can seek to prevent them.

Chap-ter 9 provides a brief outline of a number of documents – codes, guidelines and legislation – which are of relevance to research, and with which the researcher should be familiar. The report ends with a bibliography.

To give more tangible substance to our account, we have supplemented it with a number of examples from the day-to-day life of the researcher. These examples are fictitious, but nevertheless realistic. One reason for including them is to show that, in practice, professional ethics in the field of research can entail difficult choices between different courses of action. The ques-tion is how the researcher is to act in a complex reality, in which conflicting principles have to be considered.

The focus of this account is on the individual researcher and research group. We appreciate that the ethically challenging situations that may con-front a researcher often have structural rather than personal causes. Indeed,

(12)

they may be entirely a result of conditions in the wider society, far beyond the researcher’s own control. But, even in such cases, researchers may be forced to choose a course of action and will thus have to assume responsibil-ity. In such situations, they may need support. Hopefully, this publication can help to provide that, above all by encouraging a discussion in the re-search community about how ethical dilemmas can be handled.

The faculties of universities and colleges have a particular responsibility to ensure that good training is made available for postgraduate students on different aspects of good research practice and research ethics. It is our hope that seminars and courses on the subject will be held at the country’s higher education establishments, and that this publication will be of use as recom-mended reading in that context. Our report is also intended to serve as a basis for discussion when an allegation of misconduct or a deviation from good research practice needs to be addressed.

(13)

(14)

2 Good ReseaRch PRactice – soMe

Basic RequiReMeNts

2.1 the value of research

The scope of research is vast. It extends from the smallest component parts of matter to the overall structure of the universe. It takes in the diversity of biological life, and humans as biological, culturally creative and social beings. It concerns past, present – and future.

Scholarly research is an important facet of the life of our society. The value of new knowledge is emphasized in many different contexts. What is it that gives research its value? One answer is that its value is dependent on the benefits and advantages which the new knowledge produced can bring for individuals and society. Many technical innovations that have made our lives simpler and physically less demanding spring from research in the nat-ural sciences. The results of medical research have been a major factor in reducing premature mortality, alleviating suffering and enhancing indivi-dual quality of life. Diseases that were fatal only a few decades ago can now be treated and cured. Knowledge about the human mind, historical processes and other cultures helps us to understand various aspects of modern-day society and to handle problems arising there. And so the list could go on. Indeed, research is often undertaken with the direct aim of finding an application or solving a specific practical problem.

But scientific research and knowledge are of value not only as an

instru-ment, i.e. a means of achieving something else by which we set store. The

knowledge gained is also of value in its own right, regardless of how it might

conceivably be used. We humans wish to see connections and arrive at ex-planations and understanding, even when we are not specifically looking for a use or application. Often this is put forward as an argument for basic research. Later, the results may also turn out to be of value in promoting things that we perceive to be useful and beneficial to society. Because of the very nature of research, however, we cannot know in advance exactly where its results may lead. In many cases, our desire to know and to understand is sufficient justification for engaging in it.

A third reason for undertaking research emphasizes the process, rather

than the results. A search for new truths and outlooks can be of value in itself, for individuals or for a culture or society, even if clear-cut results are never achieved.

(15)

2.2 demands on the researcher

Research, then, holds an important position in modern society, and much is expected of it. But that also turns the spotlight on the researchers, the people who are to provide the new knowledge desired. Researchers have a responsibility towards society, which in many cases also funds and commis-sions their work, and towards other researchers – and are seen as represen-ting the values for which research stands. Society – both the general public and other researchers – must be able to rely on their having done their best to arrive at valid and relevant results. There is also an expectation that re-search and those who conduct it will be free from outside influences and manipulation, and that individual researchers will not pursue their own private interests or those of other parties, especially not in an undisclosed or unconscious manner. A first step in that direction is for the researcher to try to understand what his or her own ties and motives actually are.

The standards of conduct which a researcher is expected to observe are linked to the role of the researcher, as it is understood today, and are inher-ent in the actual research process. And yet they are rooted in the ordinary ethical standards and values of society. One example touched on above is the requirement of honesty. If a researcher fails to live up to this, it may mean that what he or she is doing can no longer be called research (at least not by contemporary society), and will be regarded as something else.

Modern-day research requires high standards of both quality and integrity on the part of the researcher. Carefully considered ethical attitudes and conduct are therefore called for in the various roles which the researcher assumes.

2.3 Research areas, disciplines and environments

Good research practice is something that can be required of research of every kind. It is applicable to every discipline and field of research, both in the natural sciences, engineering and medicine and in the social sciences and humanities.

Different research areas and disciplines can be defined in terms of the different types of questions they seek to answer, the different methods they employ, the different theories on which they rest, or the different kinds of data which are available and can be collected. We can also distinguish be-tween research that is exploratory and creates hypotheses and research that tests hypotheses. In addition, research may be concerned with describing,

(16)

analysing or interpreting a body of material. Some questions can be an- swered by means of quantitative studies, while others are best addressed on the basis of qualitative studies and methods.

The environments in which research is conducted and researchers are trained can also vary widely in character and thus give rise to many types of ethical problems. In the social sciences and humanities, research is often a single-handed venture. In engineering, natural sciences and medicine, on the other hand, research teams and groups are common. Sometimes research can also be undertaken by larger consortia. Within a team or group, individual responsibilities may at times be unclear; here, collaboration is required, en-tailing various responsibilities towards the other members. At the same time, individual group members need to have a right to data of their own and to pursue lines of inquiry of their own. One research environment may be organized according to a strict hierarchy and hedged around with rules. Another, representing the opposite extreme, may be bohemian, chaotic and more or less unregulated. Obviously, various intermediate forms will also be found.

2. the quality of research

Despite the considerable variation between different areas of research in terms of theories of science, traditions and ways of working, we would argue that questions concerning good research practice can be discussed in a mean- ingful manner across disciplinary and faculty boundaries. The demand for quality in research can be broken down into a number of general principles that are widely recognized within the research community.

First of all, the basic assumptions and points of departure of a study must be made clear and explained. The project should have a clear aim in terms of answering or shedding light on certain interesting questions, which should also be stated. The researcher must be able to explain the methods to be used, and should be able to show that, by these methods, answers to the ques-tions posed may be found. The methods must be deployed in a correct and competent manner, and empirically based projects should involve a system-atic and critical analysis of carefully collected data. Arguments should be clearly formulated and relevant to the conclusion the researcher wishes to reach. The project as a whole, its documentation and the report setting out its results should be characterized by clarity, order and structure. But quality also includes things like scientific imagination and originality. If a project is creative and innovative in some respect, its quality will be significantly enhanced.

(17)

The list of requirements given here is of course by no means exhaustive. Nor can each of the criteria on its own be seen as a sine qua non for the qual-ity of a project. There must, for example, be scope for exploratory studies without clear goals. But a project that is lacking in several of the respects referred to above will often be considered to be lacking in quality.

2.5 honesty and integrity

A demand that can be made of all research is that, when analysing, interpret-ing and presentinterpret-ing their own results, or citinterpret-ing those of others, researchers never distort or embellish those results in order to gain support for their hypotheses. Nor should researchers restrict themselves to citing research or data that corroborate the hypothesis they wish to pursue. Evidence to the contrary – if known – should also be presented.

What stakeholders the results may possibly benefit or damage is, in this regard, irrelevant. Researchers have to be able to draw whatever conclusions their data support, even if from certain points of view they are undesirable. As far as possible, therefore, they must be free in their work from any ties that could put at risk the possibility of achieving such knowledge as their research is intended to produce.

This respect for the results obtained, or for “the truth”, can be expressed here as a demand for integrity on the part of the researcher. Researchers should seek to adopt a critical stance towards their own expectations, and those of others, about what the data will demonstrate, and to their own and others’ hopes of ground-breaking conclusions or rapid career progress. Shortcomings in integrity, too, affect the quality of a research project.

Nor should a researcher pass over in silence earlier investigators who have put forward the same or similar ideas to those now being tested. For the researcher, the requirement of honesty is very far-reaching. Research, by its very nature, involves a search for new knowledge, for an understanding that is as well-founded as possible – and researchers demonstrate their honesty precisely by respecting the results they arrive at.

2.6 Merton’s cudos norms

In the 1940s, the American sociologist Robert Merton formulated four prin-ciples which he believed constituted a “moral consensus” in science.

(18)

monly referred to as the CUDOS norms or rules, they have been of great significance for both the substance and the historical development of the discussion about the ethics of research. Several scholars and other contribu-tors to the debate have subsequently developed on or modified these prin-ciples, among them the physiologist André F. Cournand with his “Frensham Formulation”. Merton’s norms merit attention as one starting point for a discussion about what constitutes good research practice.

The norm of communism, or communalism (C), means that the research

community and society as a whole have the right to be informed of the results of research. New knowledge should not be kept secret and con- cealed for non-scientific reasons. According to Merton, therefore, there is no such thing as “intellectual property”, owned by the researcher. His norm of universalism (U) requires scientific work to be evaluated with reference to scientific criteria alone. When assessing the validity of the results, we are to take no account, for example, of the researcher’s race or position in society.

The norm of disinterestedness (D) means that the researcher must have no

other motive for his or her research than a desire to contribute new

know-ledge. The fourth norm, organized scepticism (OS), requires the researcher to

constantly question and scrutinize, but also to refrain from expressing an assessment until he or she has sufficient evidence on which to base it.

Since these principles were put forward, the position of the researcher, or at least the general perception of it, has changed in many respects. Be-ing a researcher can no doubt colour an individual’s whole way of beBe-ing and thinking, but these days it is quite a common professional role, and researchers are employed specifically as researchers. They, too, are expected to be loyal to organizations and superiors, and have to take financial factors and their own job security into account. In many cases, therefore, Merton’s

norms will be difficult to live up to in reality. His rule of disinterestedness,

which says that the researcher’s main reason for doing research should be to contribute new knowledge, is a case in point. Researchers must surely be allowed to have other motives as well, such as promoting their prospects of employment through the work they do. The important thing, rather, is that motives of this kind do not influence the researcher in such a way that he or she arrives at interpretations or conclusions for which there is no scientific basis, or withholds findings for which evidence does exist.

Merton’s strict requirement of communism is also difficult to live up to in

many types of research and in certain research environments, for example in an industrial setting, although the importance of publishing results and communicating them to society and to other researchers will neverthe-less often be acknowledged in such environments as well. In addition, this norm may in certain cases have to take second place to other ethical re-

(19)

quirements, such as the protection of informants. There are various pro-blems with Merton’s other norms, too. The ideals expressed in the CUDOS norms nevertheless provide one of the cornerstones for the present-day di-scussion about research misconduct (see chapter 8). They are also reflected in the requirements of honesty and openness that were formulated in our introduction (see chapter 1).

2.7 other roles of the researcher

So far we have discussed the requirements of quality and integrity chiefly with regard to actual research, but they also need to be discussed in relation to the other activities which a researcher may undertake. Researchers are often called upon to make assessments, for example of the work of colleagues, as external assessors in connection with appointments or as referees or reviewers for articles submitted to journals. They may supervise post-graduate students, and often teach, not only at higher education establish-ments, but also in other contexts. In addition, they may communicate re-search results to the public or to specific groups, for instance via the media. A researcher may also, in his or her capacity as a researcher, be appointed to various boards or committees. In all these contexts, he or she is generally perceived as a representative of the research community. In these roles too, therefore, honesty, integrity and respect for knowledge need to characterize the researcher’s actions.

In chapter 7 we will discuss what good research practice can entail more specifically in relation to these other roles of the researcher. That discussion, like the rest of this report, will be expressed largely in general terms, rather than focusing on any particular field of research. However, it should be easy for anyone with personal experience of research to relate what is said to relevant, concrete situations. In some cases, more detailed examples will be provided.

One of the responsibilities of researchers working at universities and col-leges is to disseminate their results and thus contribute to the development of society, and this is something we should also be able to demand of those employed in other organizations. In addition, researchers are sometimes expected to make an active contribution to public debate about social or political issues, for example in the media. In such contexts, good research practice requires the researcher to critically consider what course of action is appropriate, and to be aware of the risks that can arise if the confidence placed in the expertise of the researcher is misused to support positions that fall outside his or her sphere of competence.

(20)

2.8 Research ethics and professional ethics

In this report we primarily discuss questions concerning the quality of search, the honesty and integrity of the researcher, and the conduct of the re-searcher in certain roles other than actual research. We do not, in other words,

examine what is often considered the primary focus of research ethics (and what

is commonly meant by the Swedish term forskningsetik), i.e. issues relating to

how researchers take into account and protect the interests of participants, informants, subjects and others affected by their research. Research ethics in that sense is concerned, for example, with how human subjects are selected, what information is provided and how, how participation affects participants during the project and on its completion, and how publication may affect participants, as well as with questions relating to third parties, for instance regarding information, publication and how such parties may be affected.

In Swedish, the word forskaretik, which could be translated as the

profes-sional ethics of the researcher, has been suggested as a term for the area dealt with in this publication. The emphasis here is on the researcher’s ship to the actual role and task of research, rather than his or her relation-ship to participants or third parties. The term does, though, encompass the researcher’s relationship to co-workers, other fellow researchers and fund-ing agencies. The distinction between the two concepts is difficult to define satisfactorily, however.

It is also possible to distinguish between internal and external research

ethics. Sometimes this distinction is used in a way that coincides with the dis-tinction between the professional ethics of the researcher and research ethics in the sense outlined above, but sometimes it is given a different meaning.

The following example describes what is primarily a problem of research ethics in the sense of issues relating to participants and others affected by research.

What would you do in the following situation?

A researcher discovers that a famous writer had an original and unusual sex life, and that his marriage was a mere façade to hide it from the outside world. The researcher relates this dis-covery to some disputed passages in the writer’s work, arguing that it sheds new and original light on them. However, the widow and several other relatives of the writer are still alive. They are offended by the researcher’s publication of his findings, and contest his claim.

What consideration should the researcher show for the feelings of relatives in a case like this? Would it have been different, in your view, if the subject of the research had been a well-known politician who had vigorously championed the cause of sexual minorities? What do you feel about the differing guidelines that apply to researchers and journalists in such situations?

(21)

In postgraduate training, attention obviously needs to be paid both to the professional ethics of the researcher and to research ethics, in the senses used here. A familiarity with the legislation, guidelines and ethical codes that apply to a researcher’s particular field is desirable and important. Such documents include, in Sweden, the regulations of the National Archives, the Personal Data Act, the Act concerning the Ethical Review of Research Involving Humans, the regulations of the Medical Products Agency, the Animal Welfare Act and Animal Welfare Ordinance, the Declaration of Helsinki, the Swedish Research Council’s guidelines on medical research

(Riktlinjer för god medicinsk forskning) and ethical principles for research in

the humanities and social sciences (Forskningsetiska principer inom

humanis-tisk-samhällsvetenskaplig forskning), and the various rules and conventions on research adopted by the European Union (EU) and the Council of Eur-ope. Some of these texts are described in chapter 9.

The ethical issues which a researcher may need to consider are not, how-ever, restricted to the areas referred to above as research ethics and profes-sional ethics, as the examples below make clear.

What would you do in the following situations?

You have been given permission by the National Prison and Probation Administration to study the reading habits of prison inmates. You discover two shocking cases of physical abuse in the prisons you visit, but both the victims ask you not to raise the matter with the prison management, as they are afraid of reprisals.

Do you as a researcher have any special responsibilities in this context?

Researchers working in certain fields can sometimes, through their research, gain insight into situations, for example in people’s homes or at day nurseries, which raise ethical questions. How should you handle the conflict that can arise when, as a researcher, you come into contact with instances of neglect and abuse? On the one hand, trust has been placed in you, for example by your being admitted into a family; on the other, as a profes-sional you have a responsibility to report ill treatment of elderly people, children etc.

What would you do in such a situation?

In a research project involving analysis of tissue samples, one sample reveals a type of cancer for which the prognosis is very good if the disease is diagnosed early. By mistake, the tissue sample has not been de-identified.

What do you do with this information? Do you contact the person concerned or not?

(22)

(23)

(24)

3 PlaNNiNG ReseaRch

We will now discuss various aspects of what good research practice may en-tail, and will do so by following, in broad outline, the different stages in the research process. This chapter will chiefly be concerned with the reasons for undertaking research and the planning of a research project. In the next two chapters, the focus will be on how research is conducted and how the results are handled.

3.1 the purpose of the research

In the introduction we argued that, from a general, societal standpoint, re-search can be undertaken with a view to achieving beneficial and useful applications, but that it can also have the important and more fundamental aim of increasing our stock of knowledge, and thus broadening our per-spectives and contributing to a richer life. For the individual researcher, the purpose may be more personal, such as satisfying a sense of curiosity, an enjoyment of problem-solving or a mania for collecting, indulging a love of fault-finding and criticism, helping to tackle problems in society, impress-ing other people, advancimpress-ing one’s career, or earnimpress-ing money from inventions and patents. This is not the place for a closer scrutiny of such personal mo-tives, which in many cases are probably mixed and perhaps not consciously apparent to the researcher concerned. But they may influence the direction of the research undertaken, and the environment in which it is carried out.

The research community has traditionally taken a generous view of re-searchers’ personal motives. That is a good thing, in that it allows many people, with varying motives, to contribute to the research culture and the growth of knowledge. On the other hand, a research environment which for example emphasizes individual careers may become destructively com-petitive. It may make it impossible for other researchers, not motivated pri-marily by a concern for their careers, to assert themselves or feel at home. And an environment focused on the personal gains to be made from patents could result in a distorted emphasis on short-term benefits. A research set-ting with a one-sided focus on the value of basic research in its own right, by contrast, may become an “ivory tower”, disdaining any attempt to see research and its possible applications in a social context.

These risks will be reduced if the approach to researchers’ motives re-mains open and generous, not least within the research community and

(25)

individual research groups. The important thing should be that someone

wishes to contribute to the research endeavour, not why they wish to do

so. However, in our opinion an open discussion of personal reasons for doing research, within research teams, departments and faculties, is also im-portant. An awareness of one’s own motivation must be seen as one of the outcomes of a good researcher training programme. Among other things, such an awareness should help to ensure that more applicants for permanent research positions maintain their commitment to research over the years. If that commitment sometimes falters in researchers who have finally secured such posts, it may be because they have for a long time been unclear about their reasons for doing research.

It must surely be reasonable to argue that a researcher ought to believe in the value of what he or she is doing in a project. This is important on several counts. Researchers who no longer believe in their own work can eas-ily generate a stultifying atmosphere in their department, not least among younger colleagues and students. Such researchers need to rethink what they are doing. At best, such a reappraisal will result in their turning their attention to problems that they find more inspiring, and which are per-haps more important for society as a whole. Departments should encourage, rather than stifle, an ongoing discussion about the relevance of research projects in various respects, and researchers wishing to pursue new lines of inquiry must be supported by colleagues and research funders.

Another important issue for the individual researcher is the choice of re-search problem. He or she may for example be faced with a choice between a well-defined problem which could relatively quickly lead to publishable results, but which does not seem of very great relevance to society, and one that is vaguer or less beneficial in career terms, but of major social signifi-cance. The choice has to be made by the individual researcher, and resear-chers who decide to change direction and leave a project on ethical grounds should be respected by colleagues and the research community as a whole.

Such a step is not easy to take, however, especially for researchers in the early stages of their careers. But both the research community and society at large stand to benefit significantly from enabling researchers in this situ-ation to pursue new directions. Funding agencies need to bear this in mind. There is a danger of such “defectors” being if anything penalized by colleagues in the peer review process, who may concentrate more on past per-formance than on potential when assessing the new projects they propose. At worst, such narrow assessments may be the result of a blinkered outlook on the part of senior reviewers, hardened by a life of hard work and sacrifice in pursuit of their own careers.

(26)

A problem of particular importance arises when researchers realize that they are involved in research which they themselves judge could have det-rimental consequences. Such an assessment is often very difficult to make, but even so the individual researcher concerned is frequently the person in society who is best placed to attempt it. Situations of this kind were discussed in conjunction with the drafting of the Uppsala Code of Ethics for Scientists, published in 1984 (see Rydén under references). This code recommends that, as far as they are able, researchers should try to arrive at such judgements, and that if they conclude that the applications of their research will cause more harm than good, they should discontinue the work and publish their assessments. The code also calls on colleagues and the research community at large to support such individuals. We concur with these recommendations of the Uppsala Code.

Sustained individual commitment to a research undertaking is also im-portant, however. Most research projects are of such a character that they require a very substantial investment of effort and concentration. The jour-ney from initial idea to completion is as a rule both long and uncertain. Most research involves clearly creative elements, but in between there are frequently long, arduous and uneventful stretches of road. Meanwhile, oth-ers are often dependent on the researcher maintaining a certain momentum. Devotion to the task, application and an ability to concentrate are therefore important qualities in a researcher and a research environment. What is more, a researcher who has undertaken to present a doctoral thesis on a sub-ject or to carry out a prosub-ject, after receiving support from a funding body, should not simply abandon the task because he or she no longer “feels like it”. That is not the same thing as wanting to change the direction of one’s research on ethical grounds. There may be different reasons, then, for a re-searcher to withdraw from a project already embarked upon, and they are not all equally good.

A capable and productive researcher tells you at a party that he no longer enjoys doing research. Asked why he continues all the same, he says: “I can’t do anything else, and I have to make a living and pay maintenance for three children. But I never have time to see them. I’m not sure I can go on much longer.” A few days later, a gifted student whose final-year thesis you are supervising says she is thinking of going on to do a PhD and is keen to have your colleague as her supervisor, “because he’s so incredibly enthusiastic”.

Do you do anything?

(27)

3.2 describing and documenting the project

Really good research can rarely be planned in detail. Often its focus has to change significantly as the work progresses. This is inherent in the nature of research. It is, after all, a matter of exploring the unknown, and that will bring surprises for those open and flexible enough to see them.

That does not mean that it is not a useful habit to try to describe in ad-vance what you intend to do – to draw up and follow a project plan, while still being free to modify it on an ongoing basis. This is particularly important when doing research in a group, and when several young researchers and PhD students are involved. They are appointed on fixed-term contracts and often have to report results at frequent intervals to secure renewal of their contracts. They need such a plan for their own planning, and to give their work the necessary long-term stability. The project plan should also indicate who does what and, if possible, include estimates of the time needed for dif-ferent stages in the work. What is more, a research plan is required by most research funders, and by regional ethical review boards. And detailed plans are particularly necessary when human subjects are involved, for example in clinical trials of drugs or large-scale treatment studies.

It is also a good idea to keep a regular record of how the work progresses. Especially in a group setting, continuous, structured documentation of this kind is of great importance. Researchers sometimes have occasion, later in life, to deeply regret not having properly documented some aspect of a study. If there is no documentation and no one can remember exactly what was done, it is impossible to give clear answers about details of the work, either within the group or to others. Going back to try to reconstruct what happened involves a duplication of effort that could have been avoided. Regular documentation is of course particularly crucial in an ethically sensitive project. An additional reason for it is that, without proper records and retention of source data, it can be more difficult to respond to allegations of research misconduct and to clear a researcher’s name (cf. chapter 8).

3.3 applying for funding

There are a great many sources of research funding in Sweden, including foun-dations and organizations of various kinds and central government agencies. Major bodies which distribute public funds include, for example, the Bank of Sweden Tercentenary Foundation and the Swedish Research Council.

(28)

Grant-giving bodies expect an application for funding to provide a clear description of the intended project, set out under a number of specified headings, as a basis for assessing the scientific quality and potential signifi-cance of the project. Usually this presentation will include an account of the current state of research (the scientific background), a formulation of the problem or problems, and a description of the purpose of the research. Re-searchers are also asked to state what data they intend to gather and how they will go about it, what methods and approaches they plan to use to analyse the data collected and, where relevant, what hypotheses they intend to test.

In addition, the application should give the names of the principal investi-gator and the other researchers to be involved, as well as describing the rele-vant experience and qualifications of each participant. As a rule, applications also have to state whether a similar application is being made to other fund-ing bodies, with the same or another principal investigator. Any commercial or comparable interests and affiliations which could affect the project should also be declared; for funding from the Swedish Research Council, for example, such information is obligatory. Relevant factors here include grants, employment and consultancy fees, and also any links between the project and business activities engaged in by the researcher. If funding is being provided from another source, the application has to state what conditions apply and how the implementation and results of the project could be affected (see 9.8 below concerning the Swedish Research Council’s guidelines).

Recipients of research grants are expected to submit regular progress re-ports, describing how the funds have been spent and what has been achieved during the period covered. The scientific reports submitted in this context should give a clear account of the work for which the grant was sought and has been used during the funding period.

Applicants sometimes feel that too much red tape is involved, in terms of information to be supplied and forms to be correctly completed. It has to be remembered, though, that funding bodies receive hundreds of applica-tions every year, and that to a large extent these applicaapplica-tions are evaluated by fellow researchers, often to tight deadlines, in their spare time and for little financial reward. It is therefore important to ensure that proposals are relatively easy to compare, allowing reviewers to focus on assessing their scientific value.

In the light of our own experience, we would recommend proposal writers to be as honest and open as possible. Submitting an application which can-didly describes sources of error, makes a balanced appraisal of the project’s value etc. usually pays. To try, as many do, to “sell” a project by exaggerating its significance is a deplorable practice, and often counterproductive. What is more, here as in other scientific contexts, exaggeration tends to discredit

(29)

those who practise it. It also contributes to a culture which – not least in the applicant researcher and group themselves – undermines a proper sense of moderation and balanced criticism, qualities that are of great value in research training and research itself and therefore need to be encouraged in the research environment.

Nor should the title or presentation of a project be slanted to give the false impression that the proposed research lies within a field given priority by the funding body. Twisting the truth in this way is common in funding applications and tends in turn to engender a fairly brutal cynicism on the part of the agencies receiving them. As a researcher – if only to retain the respect of funders – one should avoid perpetuating this bad habit.

Equally, an honest application should not have a hidden agenda. Seeking a grant, for example, not because you wish to carry out the project, but to reassure a staff member by pretending to try to secure research funds for him or her, and calmly reckoning on the application probably being turned down, or even letting the reviewers know, off the record, that funding is not all that important, is obviously a practice that should not occur.

Likewise, no attempt should be made to mislead reviewers by tactically including an applicant, or even a main applicant, who it is believed will enhance the status of the application, if the person concerned is not in fact prepared to accept full responsibility for the project. This type of deception usually backfires in several ways. Many reviewers routinely check what con-tributions the different applicants can realistically be expected to make and, with fierce competition for the funding available, this can provide them with a good reason to reject a proposal. What is more, heated disputes can easily arise over ownership of a project and its results if a grant is awarded and a nominal applicant begins to make unanticipated claims.

Another form of dishonesty is to exaggerate in the application what resources are required, on the basis that the sums will in any case be reduced and, this way, funding may end up at an acceptable level. Such exaggerations also give rise to quite a brutal culture among reviewers, in the long run devaluing the credibility of researchers. In addition, they increase the risk of honestly estimated requests for funding being reduced as a matter of course.

It is of course also important to be honest in one’s assessment of the potential ethical problems of a project and how they can be managed or resolved. Since it is only recently that such issues have gained wider attention, applications are still received in which researchers fail to adequately address them. Careful ethical assessments, however, are a mark of quality, not only in research reports, but also in applications. A well-considered examination of ethical concerns may very well be a deciding factor in the evaluation of a proposal.

(30)

If a researcher wishes to use a research grant for other purposes than those for which it was originally awarded, he or she should inform the funding body and enter into negotiations with it in good time. In the case of com-missioned research, special conditions may apply. Minor deviations from the original plan are natural, and usually acceptable. But since it is difficult to draw clear dividing lines in this regard, it is always advisable to contact the grant giver when you know that you have made significant changes. It is also important to bear in mind that a modification of a project could raise new ethical problems. Sometimes this may necessitate a new ethical review of the project.

A special form exists for applications for appraisal by a regional ethical re-view board, the various points on which are together intended to illuminate the ethical aspects of the project. An application that has been submitted to a public authority, such as an ethical review board or the Swedish Research Council, becomes an official document from the moment it is received. If there are no grounds for classifying it as confidential, it also becomes a pub-lic document. Its contents are thus available to anyone who requests access to it, which can cause difficulties for example in connection with patent applications. Ways of solving these problems are currently being discussed.

(31)

(32)

coNductiNG ReseaRch

.1 choice of method

The method chosen for a research project is crucial to the value and char-acter of the results. The choice is often difficult, requiring considerable experience, and often boldness. Not uncommonly, it is made on the basis of knowledge acquired and work undertaken in the past, perhaps by earl-ier generations in the same research group or department. Sometimes the research environment at the department in question is so “steeped” in this methodology that alternative approaches are simply not discussed or even considered. In such cases it can be useful to deliberately look for alternatives and perhaps (possibly in collaboration with researchers from other metho-dological traditions) to undertake parallel studies using different methods.

In scholarly research, questions of method are crucial, and linked to cri-teria of scientific quality. This is true not least in the humanities and so-cial sciences. There is, for example, a difference that is more than merely practical between studies of humans based on measurements, say, of reac-tion times or response frequencies in structured quesreac-tionnaire surveys, i.e. “positivist” methods, and “hermeneutic” studies in which the researcher interprets the thinking of individuals, as revealed for instance in correspon-dence or in-depth interviews. In a discussion of what method to choose, the generalizability and supposedly more or less objective character of the results may have to be weighed against the interest and “depth” of the scientific conclusions that can be arrived at. This does not mean, though, that it cannot be particularly rewarding to attempt collaborative research which combines positivist and hermeneutic methods. The choice between these two approaches also has ethical implications. In studies of the first- mentioned kind, the researcher’s relationship to the individuals studied is often cooler and more distanced; in the second type, it involves a greater element of empathy. In both cases, the researcher’s position can entail ethical complications or risks.

Choices of method can also involve other important ethical deliberations. It may be a matter of deciding whether animal experiments can be com-pletely or partly replaced with work on tissue samples, whether a drug trial should be abandoned when serious adverse reactions emerge, or whether placebo treatment of a control group should be cut short when the drug is found to be effective. It may also be a question of how to limit the scope

(33)

of an interview study of children of abused mothers, to what extent vio-lent tendencies or intelligence should be measured in socialization studies of different ethnic groups, and so on. In principle, the ethical aspects of such work should be considered by a regional ethical review board. Gener-ally speaking, for a particular piece of research, a method should be chosen which minimizes the conceivable harmful consequences for the individuals concerned, if in other respects the methods available are of reasonably simi-lar merit. In addition, the benefits of the research planned, and the scientific value of the results that can be expected, should be weighed against any adverse impacts.

In the natural sciences, medicine and other fields, in Sweden and elsewhe-re, the work of research groups tends to be quite strongly method-driven, being based on a methodology developed within the group and forming a unifying link between a number of different projects in which it is em-ployed. In such cases, the choice of research problem may be determined by the preferred method. This is at odds with the conventional notion of the researcher as a problem solver, who first asks a question and then chooses a method to answer it. At worst, the research undertaken by a method-based group will become fragmented, and much of it may be quite superficial. On the other hand, a systematic exploration of the potential of a newly devel-oped method can be of great value. It also has to be said that, in general, many advances in the modern natural sciences, from astronomy to brain research, are to be regarded largely as a result of developments in technology that have made possible new methods. Advocates of “pure science” may prefer to ignore it, but the fact remains that technical progress has made possible many of the advances in the natural sciences, just as scientific discoveries have made possible progress in technology. Research groups working in nat-ural sciences have every reason to acknowledge and value the contributions made by technical specialists, both within and beyond such groups.

A PhD student has a supervisor who received her own training in a positivist tradition and who has a solid grasp of quantitative methods and concepts. The student is working on a problem that could also be studied using other, qualitative methods. He wants to study these methods more closely, and has the support of others at the department. But his supervisor wants him to stick to his original research plan, and threatens to withdraw as his supervisor if he does not. The PhD student realizes that, if he insists, it could dam-age his career.

What should he do?

(34)

.2 data handling and archiving

The various data and samples that are collected in the course of a research project are referred to as source, or primary, data. They may consist of in-strument output from experiments, astronomical images from space tele-scopes, X-ray images from hospitals, tape recordings of interviews, results of large-scale computer simulations, soil samples, tissue samples and so on. Many researchers put a huge amount of effort into gathering such material. At many departments, researchers regard their source data as their personal property. That view, though, has to be questioned. The work of collecting the data has as a rule been done as part of a contract of employment, often with public funding; society may thus have invested major sums in the study in question. The value of material of this kind may be considerable, for later generations as well, and perhaps for reasons which cannot currently be foreseen.

It is important, therefore, to handle source data with great care and to retain and archive them in such a way that they can be made available to researchers other than the ones who gathered them. In the shorter term, this is important in order to allow published results to be verified following publication, for example to trace sources of error or to enable the researcher to counter accusations of research fraud. Investigators therefore cannot promise that no other researchers outside the group collecting the source data will ever, under any circumstances, be given access to those data. A PhD examiner, for example, may need to be able to review the data collected.

Establishing procedures for archiving can hardly be regarded as the re-sponsibility of the individual researcher or research team alone. Such pro-cedures have to be put in place and maintained by departmental managers. But in practice responsibility for ensuring that they are followed often falls on the project leader and individual researchers.

Obviously, every piece of data, every computer run, every test result from an accelerator etc. does not have to be preserved for posterity. As part of its procedures, a department should have ways of determining practical limits to the endeavour to document research. But it is important to make sure the procedures adopted are in fact applied by all concerned. Nowadays, the turnover of postgraduates and young researchers holding short-term posts is so high at many departments that large quantities of data and the know-ledge associated with them are regularly lost by research groups because adequate procedures for documentation and archiving either are not in place or are not followed.

(35)

The research council is short of funds this year, and you have to make cuts in your pro-jects. At the same time, one of your PhD students is on parental leave, and another has just finished and left the department. You are in the middle of a major data collection project. Your own processing of the data is in danger of making no headway, as it is so time-consuming. Archiving the data is not straightforward either. You could do with set-ting up a new digitized archiving system, but while this will have to be done in the longer term, now is not the right time. In addition, the doctoral student who has just left did not keep very careful records of his initial processing of the data, with the result that the task of archiving them feels insurmountable. Directives have been issued centrally, requiring data to be archived and made available. This threatens to bring the project to a complete standstill for at least six months, preventing you from publishing and obtaining continued funding.

What do you do?

In Sweden, processing of data and archiving are subject to legislation which also applies to research (see section 9.12).

A researcher, Adam, collects data from a specific group of adult informants. He promises that no one outside his research group will have access to the data. Later his findings are questioned by two other researchers, Brian and Cecilia, who request access to his source data. Adam refuses to hand them over, referring to his promise to his informants. The case reaches an unexpected conclusion when colleagues of Adam’s say they have destroyed the source data on their own initiative.

Is the action taken by Adam’s colleagues ethically acceptable? Is it compatible with existing legislation? Has Adam promised more than he can deliver?

The above example draws attention to problems that can arise in both social science and medical research, and can help to clarify the principles mentioned earlier in this section. As regards what a researcher can promise, what information should be retained, how source data should be archived and who should have access to them, it may be helpful to distinguish between a number of different situations:

1. Verification of data, e.g. in conjunction with the presentation of a Phd thesis

In our opinion, access to source data for verification purposes is non- negotiable. That is to say, a researcher cannot promise that no other researcher, under any circumstances whatsoever, will be allowed to scrutinize the data. This is of course subject to the condition that the examiner or other

(36)

person looking at the data undertakes to observe the same confidentiality etc. as the researcher has promised the participants. In clinical trials of drugs, such a principle has long applied.

2. secondary use of data, i.e. subsequent use in other research

It is a basic principle that source data and other research material produced with government funding, for example in the form of a grant from the Swedish Research Council, are not the private property of the researcher. Under certain circumstances they may be used again, after a certain time has elapsed and subject where relevant to ethical review and other appro-priate consents. This is a way of making research more efficient, and is al-ready systematically applied in many quarters, for example with respect to data collected at major international research facilities. The researcher can, however, restrict other investigators’ access to data for such purposes, for in-stance if the data are personally sensitive and there is good reason to believe that subjects will otherwise refuse to take part. The alternative, in other words, might be no research at all. That risk has to be weighed against the gain in efficiency terms from allowing other researchers (at a later stage) to use the data for further research.

3. investigating allegations of research misconduct

If allegations of research misconduct are made, and it appears from an ini-tial consideration of the case that they cannot be dismissed as unfounded, then a more in-depth investigation has to be undertaken. This is also in the interests of the researchers accused – only by conducting such an investiga-tion is it possible to clear them of suspicion. This can be done with the help of the Swedish Research Council’s Expert Group on Research Misconduct. This group, whose members are appointed by the Council’s Board, then has to take over all the pledges of confidentiality made by the accused researchers. Scientists who destroy data to prevent anyone else gaining ac-cess to them can never at a later date be cleared of allegations of research misconduct.

This means that researchers can never promise that no other researcher, under any circumstances, will be allowed access to the data collected. It should be made clear in the information given to participants that this may happen in special circumstances and for specific purposes, for example for a scientific review of the research, but that the researchers performing such a review will undertake to honour the promises of confidentiality which the researchers collecting the source data, or the equivalent, have made to the participants.